1. Field of the Invention
The present invention relates, generally, to scientific and medical systems, apparatus and methods. More particularly, the invention relates to method and apparatus for diagnosis of conductor anomalies. Most particularly, the invention relates to a method and apparatus for diagnosis of conductor anomalies in an implantable medical device, such as an implantable cardioverter defibrillator (ICD), a pacemaker, or a neurostimulator.
2. Background Information
Anomalies of conductors in implantable medical devices constitute a major cause of morbidity. Examples of such devices include pacemakers, implantable cardioverter defibrillators (ICDs), and neurostimulators. For example, early diagnosis of ICD lead conductor anomalies is important to reduce morbidity and/or mortality from loss of pacing, inappropriate ICD shocks, and/or ineffective treatment of ventricular tachycardia or fibrillation (ventricular fibrillation).
Multilumen ICD defibrillation electrodes include both one or more high-voltage conductors and one or more pace-sense conductors. Pace-sense lead fractures commonly present as inappropriate shocks caused by oversensing of lead-related nonphysiological potentials, commonly referred to as lead “noise” signals, caused by the conductor anomalies. Functional failure of an ICD's pace-sense conductor may result in symptoms caused by loss of pacing functions for bradycardia rate support, cardiac resynchronization, or antitachycardia pacing.
Thus one major goal is high sensitivity of diagnosis: identification of lead conductor anomalies at the subclinical stage, before they present as a clinical problem. A second major goal is high specificity: A false positive provisional clinical diagnosis of lead conductor anomaly produces patient anxiety and results in potentially-avoidable diagnostic testing. A false positive confirmed clinical diagnosis results in unnecessary lead replacement, with corresponding expense and risk.
Existing technology for diagnosis of conductor anomalies in an implantable medical device is believed to have significant limitations and shortcomings.
The primary method in the prior art for monitoring pacemaker and ICD lead integrity is periodic measurement of electrical resistance, commonly referred to as “impedance monitoring.” Impedance monitoring uses single pulses. Various methods are well-known in the art. These methods provide a value of impedance close to the direct-current resistance.
In the circuit being measured, most of the resistance is at the electrode-tissue interface of the high-resistance tip electrode, and variations of up to 10% in this value are common. Each individual pace-sense conductor (for example, the conductor to the tip electrode or the ring electrode) contributes less than 10% to the measured resistance. In some ICD leads, this value is as less than as 3%. Thus even if the resistance in a single conductor doubled or tripled, the overall measured resistance will remain within the expected range. Measurements indicate that resistance exceeds the expected range until the conductor has lost most of its structural integrity. Thus resistance remains within the expected range even when only a fraction of the conductor is intact. For this reason, resistance measurements are insensitive to partial loss of conductor integrity. Further, resistance measurements have limited specificity. A single, out-of-range value may be an artifact, and marked increases can occur at the electrode-myocardial interface.
Hafelinger et al (U.S. Pat. No. 5,003,975) and Cinbis et al (U.S. Pat. No. 5,897,577) summarize some of these methods, which include measurements made directly using either a single pacing pulse or a single independent pulse used only for measuring resistance. McVenes et al (U.S. Pat. No. 5,741,311) describe use of a longer burst of alternating current at a single frequency. The purpose of these longer (about 100 ms) pulses is to drive the system to a steady-state condition that is not achieved by single, short (less than 1 ms) pacing pulses. Schuelke et al (U.S. Pat. No. 5,755,742) describe a method for measuring resistance of defibrillation electrodes by applying a test voltage applied to a different excitation current pathway. Kroll et al (U.S. Pat. No. 5,944,746) described an automated method for periodic measurement of the resistance of the high-voltage (defibrillating) coil in ICD electrodes. Gunderson et al. (U.S. Pat. No. 7,047,083) described a method and system for automated, periodic, measurements of resistance in conductors attached to an ICD or pacemaker. However, these types of “impedance monitoring,” which return values close to direct current resistance, identify lead anomalies before inappropriate shocks in only about a third of ICD patients who have conductor fractures.
A newer method for monitoring ICD lead integrity is based on the response of ICD pulse generators to electrical “noise” signals associated with lead conductor fractures. These nonphysiological signals have specific characteristics that differentiate them from true cardiac signals such as high variability and, at times, nonphysiologically-rapid rates. If these signals are of sufficient amplitude and exceed the ICD's dynamically-changing sensing threshold, the ICD oversenses them. Repetitive oversensing of nonphysiologically-short intervals may indicate lead conductor fracture even if lead resistance is normal. Gunderson et al. (U.S. Pat. No. 7,289,851) described a Lead-Integrity Alert that incorporates both ICD-based measures of oversensing based on the nonphysiologically-rapid rate of sensed signals and periodic measurements of resistance. This method, combined with automatic ICD reprogramming, improves warning time before inappropriate shocks caused by lead-related oversensing. Nevertheless, approximately 25% of patients receive less than 3 days of warning, and some receive almost no warning.
This method detects only some lead-noise signals. It cannot detect a lead anomaly unless it generates signals that are both fast enough and of sufficient amplitude to be classified as nonphysiological oversensing. Thus it will not detect a lead anomaly if it does not generate “noise signals” or it generates only low-amplitude noise signals, or signals that do not occur at a fast enough rate.
Gunderson et al. (U.S. Pat. No. 7,369,893) further describe a method for withholding delivery of ICD shocks if ventricular fibrillation is detected from analysis of the pace-sense lead, but not confirmed by analysis of the high-voltage lead. Although not yet evaluated in patients, this method is expected to further reduce unnecessary shocks. However, it requires sufficient oversensing to result in inappropriate detection of ventricular fibrillation clinically. Thus it does not provide early diagnosis of conductor anomalies. Withholding shocks for ventricular fibrillation detected on the near-field electrogram has an inherent risk of withholding life-saving therapy, however small, and is thus not the preferred approach to diagnosis of conductor fracture. Like the Lead-Integrity Alert, it is not applicable to intraoperative diagnosis or to pacemakers and neurostimulators.
Additionally, no presently-used method reliably warns before loss of pacing function for bradycardia pacing support, antitachycardia pacing, or cardiac resynchronization pacing.
In addition to limited sensitivity, present methods for diagnosing lead conductor anomalies have limited specificity resulting in false positive diagnostics. Evaluation of false positive diagnostics adds cost and work to medical care and may contribute to patient anxiety. If a false-positive diagnostic is not diagnosed correctly, patients may be subject to unnecessary surgical lead replacement with its corresponding risks. In the only report on this subject, 23% of leads extracted for the clinical diagnosis of lead fracture tested normally after explant.
Any clinical method for detecting conductor anomalies in implanted leads must make measurements while the conductor and lead are in the body. Typically, the measuring circuit includes the conductor-tissue interface in the body. Thus the measured values will depend both on the behavior of the conductor being evaluated and the conductor-tissue interface. Warburg (1899) was one of the first to study frequency-dependent properties of the electrode-electrolyte interface. Geddes (1972) reviewed this subject extensively and (1971) studied the frequency response of stainless steel electrodes used for research. De Boer and van Oosterom (1976) studied the frequency response of platinum electrodes used in research. In their appendix, they derive the time course of voltage and current response to a transient input signal as a function of the frequency-dependent properties of the equivalent circuit. These works are incorporated by reference herein.
Circuits for measuring impedance at varying test frequencies are known in the prior art. See for example Agilent Impedance Measurement Handbook A guide to measurement technology and techniques 4th Edition or Johnson (U.S. Pat. No. 3,599,055), which are incorporated by reference herein in their entirety.
Methods for evaluating the integrity of conductors and insulators have been developed for other fields, especially the electrical power and semiconductor industries. For example, the time delay of a signal reflected from faults in power transmission lines is used to determine the distance to the fault (for example U.S. Pat. No. 4,766,549 to Schweitzer, III, et al.). Stewart et al (U.S. Patent Application 2008/0309351) describe a sensor for monitoring of high-voltage insulation (used in power generation, transmission, or distribution systems) that includes the time course (“shape”) of a response to a transient input. Bechhoefer and Sadok (U.S. Pat. No. 7,120,563) describe a method for wire-fault detection, citing as applications the aircraft and aviation industries. In their method, signals from a wire are analyzed to determine if they are characteristic of an intact wire or a faulty wire. Kwon et al (2009) describe a method for early detection of degradation of solder joints using differences in response to direct current and radiofrequency signals.
For these and other reasons, a need exists for the present invention.